Rutgers APC RGBD Dataset
收藏资源简介:
Rutgers APC RGBD Dataset是由罗格斯大学创建的一个大型数据集,专注于仓库拣选任务中的RGBD基3D物体姿态估计挑战。该数据集包含超过10,000张深度和RGB注册图像,以及相应的地面实况数据,用于24个在亚马逊拣选挑战中使用的不同姿态和杂乱条件下的物体。数据集的创建过程涉及使用Microsoft Kinect v1 2.5D RGBD相机进行数据收集,并在不同的视角和杂乱状态下进行。该数据集主要应用于仓库自动化领域,旨在提高物体检测和姿态估计算法的准确性。
Rutgers APC RGBD Dataset is a large-scale dataset created by Rutgers University, focusing on the RGBD-based 3D object pose estimation challenge in warehouse picking tasks. This dataset contains over 10,000 registered depth and RGB images, along with corresponding ground-truth data for 24 distinct objects utilized in the Amazon Picking Challenge under various poses and cluttered conditions. Data collection for the dataset was conducted using a Microsoft Kinect v1 2.5D RGBD camera, with captures taken across different viewpoints and varying clutter levels. It is primarily applied in the field of warehouse automation, with the goal of improving the accuracy of object detection and pose estimation algorithms.

- 1A Dataset for Improved RGBD-based Object Detection and Pose Estimation for Warehouse Pick-and-Place罗格斯大学 · 2016年



